Meta-Active Learning for Node Response Prediction in Graphs
This addresses the challenge of improving prediction performance with limited and unbalanced data in graph-based tasks, such as traffic congestion forecasting, though it appears incremental as it combines existing techniques like meta-learning and active learning.
The paper tackles the problem of unbalanced observations in meta-learning for node response prediction in graphs by proposing an active learning method that selects nodes to observe, using graph convolutional neural networks for both prediction and selection, and demonstrates its effectiveness on 11 road congestion prediction tasks.
Meta-learning is an important approach to improve machine learning performance with a limited number of observations for target tasks. However, when observations are unbalancedly obtained, it is difficult to improve the performance even with meta-learning methods. In this paper, we propose an active learning method for meta-learning on node response prediction tasks in attributed graphs, where nodes to observe are selected to improve performance with as few observed nodes as possible. With the proposed method, we use models based on graph convolutional neural networks for both predicting node responses and selecting nodes, by which we can predict responses and select nodes even for graphs with unseen response variables. The response prediction model is trained by minimizing the expected test error. The node selection model is trained by maximizing the expected error reduction with reinforcement learning. We demonstrate the effectiveness of the proposed method with 11 types of road congestion prediction tasks.